Professional Documents
Culture Documents
A Patient-Specific Single Sensor Iot-Based Wearable Fall Prediction and Detection System
A Patient-Specific Single Sensor Iot-Based Wearable Fall Prediction and Detection System
fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TNSRE.2019.2911602, IEEE
Transactions on Neural Systems and Rehabilitation Engineering
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
Manuscript submitted for review on October 04, 2018. This work was W. Saadeh, S. Butt, and M. A. B. Altaf are with the Electrical Engineering
funded by the Lahore University of Management Sciences (LUMS), Lahore, Department, Lahore University of Management Sciences, Lahore 54792,
Pakistan under startup grant number STG-EED-1216. Pakistan (e-mail: wala.saadeh@lums.edu.pk).
1534-4320 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TNSRE.2019.2911602, IEEE
Transactions on Neural Systems and Rehabilitation Engineering
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
Fig. 2: Block diagram of the proposed Fall Prediction and Detection Prototype System.
Radar-based systems [22], and Microsoft Kinetics [23]. The detection algorithm. Section VI demonstrates the measurement
alternative approach is to utilize inertial wearable sensors such results and discussion. Finally, Section VII concludes the paper.
as a gyroscope and an accelerometer [24], [25] or the inertial
sensors of the mobile-phones [10], [26]-[27]. However, the II. PATIENT-SPECIFIC FALL PREDICTION AND DETECTION
gyroscope is typically not preferable due to higher power SYSTEM
consumption than the accelerometer [11]. To continuously The proposed fall prediction and detection prototype system
monitor fall events, it is more suitable to use wearable devices comprised of two parts; the data acquisition part followed by the
with detection and prediction capabilities as they can be utilized PS classification part. The block diagram of the proposed system
anywhere and not limited to certain locations [24]. is presented in Fig. 2. In the sensing part, a tri-axial
Several fall detection algorithms recognize multiple features accelerometer is used to extract the acceleration of the elderly
to differentiate between fall cases and regular activities of daily person in three orthogonal directions, the X, Y, and Z-axis, at a
living (ADL). These features include falling speed, stride time, sampling rate of 256 S/sec. The acquired accelerometer data
acceleration coordinates, posture information, inactivity periods, (Ax, Ay, and Az) along X-, Y-, and the Z-axis, respectively, are
and angular velocity [2], [25], [28]. The extracted features are then transferred via low energy Bluetooth interface (BLE (I/F))
then classified as a fall event or an ADL, using a threshold value to the classification part that is implemented on a Field-
or by combining posture information and kinematic thresholds, Programmable Gate Array (FPGA).
to generate a decision [11]. Nevertheless, the falling speed and A. Wearable Sensing Part
gait features exhibit deviation from person-to-person and varies
with aging too [29],[30]. Therefore, it is essential for a fall The system board is designed with an accelerometer sensor, an
detection system to tackle these differences for accurate fall Arduino (Nano) microcontroller, a 16-bit ADC, and Bluetooth
detection [31]. Currently, the fall detection systems attempt to (HC-05), powered by a 9V battery with an overall size of 6 cm
minimize the false positives; where some ADLs are incorrectly × 3.5 cm × 2 cm. The sensor chosen is the MPU-6050 Tri-Axial
identified as fall events, at the cost of long latency and enhanced accelerometer with an adjustable full-scale range of ±2g, ±4g,
computational cost [11], [32]. Moreover, current fall-detection ±8g, and ±16g, which can be attached comfortably to the
systems do not have a low power storage to record the fall’s patient’s thigh without distressing the person’s routine life
acceleration data [1]. activities. The upper thigh location is the preferred location for
placing the sensor since it connects the bottom kinetic chain
This paper presents a patient-specific (PS) fall prediction and
(legs and feet) to the parts responsible for keeping steadiness
detection prototype system. In this work, a single tri-axial
accelerometer sensor attached to a person’s thigh is utilized to (core and head) [6]. In addition, elder people with chances of
predict and detect the fall events and save their data for detailed high fall risk show less harmonic acceleration ratio sequence in
follow-up by the medical practitioner. The patient will be the pelvis and upper thigh [6], [33]-[34].
notified if a fall event is predicated while the occurrence of a fall The ideal wearing position of the wearable sensor is shown
event will initiate an alarming notice to the concern healthcare in Fig. 3 (a) on the subject’s thigh. The tri-axial accelerometer
providers by connecting it through the internet. The performance sensor is mounted inside the wearable part where the Y-axis of
of the proposed system is verified on 77 subjects with 600 the sensor is always in parallel with human body Y-axis.
recordings including 100 recordings of Fall/ADLs from 20 elder However, both X-axis and Z-axis of the sensor can be
subjects (aged above 65 years). misaligned from the human body X-axis and Z-axis,
respectively, due to wearing the device in a different
The structure of the paper is as follow. Section II describes
the proposed PS fall prediction and detection system. Section III angle (𝜃 ≠ 0) as depicted in Fig. 3(b). Both X-axis and Z-axis
details the feature extraction. Section IV elaborates the proposed will be rotated by the same angle θ forming two new axes X’
fall prediction algorithm, and Section V explains the fall and Z’. The rotation will lead the sensor to read acceleration in
1534-4320 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TNSRE.2019.2911602, IEEE
Transactions on Neural Systems and Rehabilitation Engineering
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
these different axes X’ and Z’. These new axes are related to and specificity based on features utilized in a recent study
ideal axes as given in both (1) and (2): conducted in [6]. The selected seven features provide the
𝑋’ = 𝑋 𝑐𝑜𝑠(𝜃) (1) highest sensitivity and specificity of 97.8% and 99.1%,
respectively. Most of these features can be also utilized for
𝑍’ = 𝑍 𝑐𝑜𝑠(𝜃) (2) assessing patients as high or low fall risk patients [6]. The
Ideally, we want 𝜃 = 0 where (𝑐𝑜𝑠 (0) = 1). For optimal adopted set of features for fall prediction includes 1) mean
functionality of the proposed system, the rotation angle θ should acceleration of X-axis(𝜇𝑥 ), 2) mean acceleration of Z-axis (𝜇𝑧 ),
be maintained less than 20̊ (𝑐𝑜𝑠 (20) = 0.939). 3) standard deviation of acceleration of Z-axis (𝜎𝑧 ), 4)
coefficient of variance of Z-axis (𝐶𝑂𝑉𝑧 ), 5) correlation
coefficient between X-axis and Z-axis (𝐶𝑂𝑅𝑅𝑥𝑧 ), 6) mean
amplitude deviation of X-axis (𝑀𝐴𝐷𝑥 ), and 7) the total sum
vector (SV). 𝐶𝑂𝑉𝑧 is defined as the ratio between the standard
deviation 𝜎𝑧 and the mean 𝜇𝑧 :
𝜎𝑧
𝐶𝑂𝑉𝑧 = (3)
𝜇𝑧
𝐶𝑂𝑉𝑧 is utilized to detect the variability in relation to the
mean of acceleration events. 𝐶𝑂𝑅𝑅𝑥𝑧 is used to assess the
strength and direction of the relationships between X-axis and
Z-axis accelerations and is defined as follows [35]:
Fig. 3: (a) Ideal wearing position for the wearable sensor where the sensor axes
are aligned with patient body axes, (b) Rotation of the wearable sensor forming
new axes in X’ and Z’ directions. 𝑛(∑ 𝑥. 𝑧) − (∑ 𝑥). (∑ 𝑧)
𝐶𝑂𝑅𝑅𝑥𝑧 = (4)
B. Fall Prediction and Detection System √( 𝑛 ∑ 𝑥 2 − (∑ 𝑥)2 ) − (𝑛 ∑ 𝑧 2 − (∑ 𝑧)2 )
The acquired acceleration data along each axis is processed
through a median filter followed by a high pass filter (HPF) to Where n is the number of samples in the window, 𝑥 and 𝑧
mitigate the in-band noise. The HPF eliminate the gravity are the filtered acceleration data in X and Z directions,
acceleration since the tri-axial accelerometer generates an output respectively.
that is a combination of the desired dynamic acceleration and 𝑀𝐴𝐷𝑥 describes the mean distance of acceleration data points
gravity acceleration [25]. The system works in two parallel about the mean [35]:
modes: a fast mode for fall prediction (FMFP) and a slow mode 1
for fall detection (SMFD). The FMFP operates on the incoming 𝑀𝐴𝐷𝑥 = ∑ |𝑥𝑖 − 𝜇𝑥 | (5)
𝑛
acceleration data at 256 S/sec while for SMFD the data are sub- Where n is the number of samples in the window, xi the
sampled to 8 S/sec. This is because, for the fall prediction, it is ith filtered acceleration sample in the X direction within the
required to regularly check (every 100 msec) for potential fall window and µ𝑥 is the mean resultant acceleration value of the
event to take timely actions while for the fall detection; it is window.
required to look at longer period to identify a fall event (3 sec in The general definition of a total sum vector (SV) magnitude
this system). If a large sampling rate to be used for fall detection
for the 3-D acceleration can be found by [25]:
too, extra overhead of unnecessary computation will be
performed which will consume more power. In each operating 𝑆𝑉 = √𝑥 2 + 𝑦 2 + 𝑧 2 (6)
mode, the feature extraction block extracts the selective features
for the fall and forms a feature vector (FV). A 4KB storage Where 𝑥, 𝑦, and 𝑧 (g) denote the filtered accelerations along X-
records the accelerometer data of a fall event for further , Y-, and Z-axis, respectively.
examination by physicians. If any fall event is predicted, it will
alarm the person to take precautionary actions while if a fall
event is detected, it will be transferred via the internet to the B. Fall Detection Features
health care providers to initiate an immediate help for the fallen The fall event of a person can be identified as an unintentional
elderly person. sudden change from the upright/straight position to the resting
or lying position. Therefore, when a person falls, the body’s
III. FEATURE EXTRACTION
acceleration suddenly increases. In the proposed SMFD
A. Fall Prediction Features algorithm, only 1 feature is utilized to detect the fall, the total
In the FMFP, the feature extraction block extracts the sum vector square (SVS) is computed by finding the square of
discriminating features from the tri-axial accelerometer data for the SV as given by [10]:
fall prediction. In order to identify the most appropriate
minimum set of features for fall prediction, several features are 𝑆𝑉𝑆 = 𝑥 2 + 𝑦 2 + 𝑧 2 (7)
examined for highest sensitivity and specificity on 600 patients’
recordings from our experiments and MobiFall dataset as The proposed definition of SVS compared to SV will expand
shown in Table I [6], [27]. The discriminating features for the the acceleration data range and simplifies defining the threshold
FMFP are carefully selected to achieve the highest sensitivity
1534-4320 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TNSRE.2019.2911602, IEEE
Transactions on Neural Systems and Rehabilitation Engineering
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
value for each subject as computing square root reduces the Receiver Operating Characteristics (ROC) of different
range of the output data [10]. machine-learning classifiers along with their Area under Curve
(AUC). ROC curves were obtained using MATLAB Statistics
Table I: List of tested features for fall prediction with corresponding sensitivity and Machine Learning Toolbox [38] by feeding 7 features for
and specificity.
pre-fall predication and looking for 3 consecutive decisions of
Features Set Sensitivity
the classifier. If all of the 3 decisions are showing pre-fall case,
Specificity
then the final output of the classifier will be also decided as a
X_MEAN, Z_MEAN, X_COV, X_SMA,
1 X_PFREQ, X_ENERGY, SV 89.3% 87.2% pre-fall case. Since the validation was based on the MobiFall
X_MEAN, Z_COV, X_SMA, X_RMS,
dataset [27], the classifier output and known outcome available
2 X_ENERGY, SV
85.5% 82.6% from the database are compared to compute the accuracy of the
Y_PFREQ, Z_COV,Y_MCR, Y_RMS,
87.0% 88.9%
system. NLSVM achieves the highest classification accuracy
3 Y_STD, Y_COV, SV compared to DT, LSVM, and KNN.
Z_COV, Z_MEAN, Z_MAD,
4 Z_STD, Z_SMA, SV
92.6% 95.4%
Z_COV, XY_CORR, X_MEAN,
5 X_MEAN, Z_STD,Z_MAD, SV
91.3% 92.6%
X_MEAN,Z_MEAN, Z_STD, Z_COV,
6 97.8% 99.1%
XZ_CORR, Z_MAD, SV
Z_COV, Z_MEAN, Z_MAD, Z_STD,
7 Z_SMA, SV
96.8% 97.5%
Z_COV, Z_MEAN, Z_MAD, Z_STD,
8 Z_SMA, SV 93.1% 96.8%
PFREQ: peak frequency SMA: signal magnitude area
STD: standard deviation MCR: mean crossing rate
CORR: correlation coefficient between two axes
1534-4320 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TNSRE.2019.2911602, IEEE
Transactions on Neural Systems and Rehabilitation Engineering
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
Where N is the number of support vectors (SVs), arg defines typically have low values before the fall. Therefore, the first base
the alteration of RBF, α is PS vector, and Xi and Xc are the FV for distinguishing a fall is to have all the instances of SVS values
and SV, respectively, during the learning process. Xi and Xc to be less than a predefined PS threshold (PS_TH). If this
are both 7-dimension vectors. N, arg, α, and Xc are PS condition is achieved, a value of ‘0’ will be stored in E1 and ‘1’
parameters identified during the learning phase. During the otherwise. The PS_TH needs to be updated for each patient.
classifying phase, the real-time inward FVs (Xi) is delivered to
B. Fall Impact: During this phase, a person will experience a
the NLSVM predictor with the formerly calculated PS
rapid falling of the body toward the ground, ending with a
parameters and then the fall event is predicted based on (9) [36].
N vertical shock on the ground. The duration of this high
∑ ( αi ∗ exp(−0.5 ∗ (‖Xi − XC ‖2 )/arg 2 )) acceleration spike is (300–500msec) [32]. This is shown as a
i=1 high amplitude in the acceleration curve with 𝑆𝑉𝑆 > 𝑃𝑆_𝑇𝐻.
>0 Pre − Fall
=( ) (9) Hence, the second base for distinguishing a fall case is a large
<0 No Fall spike exceeding the PS_TH in the second time frame with E2
of ‘1’.
In order to compare the system performance and hardware
resources single SVS feature FMFP versus the proposed 7 C. Post-Fall: In general, a person, after falling and hitting the
features FMFP, Table II is constructed. It shows the comparison ground or any solid surface, remains in an immobile situation for
of synthesized results based on CMOS 0.18um process of the a small duration (minimum of “1-sec” time frame) before
proposed seven-feature FV and single-feature FV i.e. SVS, with moving again [32]. This is identified by an intermission of the
NLSVM as prediction classifier. The single-feature FV horizontal line where SVS < PS_TH which is the third base for
implementation will reduce the area and energy utilization by identifying a fall event with E3 of ‘0’.
38.3% and 28.1%, respectively, but at the cost of deteriorated The three above conditions represent the entire FV, which is
prediction sensitivity and specificity of 73% and 81%, exploited in the proposed PS fall detection algorithm. The FV is
respectively, which is relatively below acceptable range for a a 3-bit code based on the result of each frame [𝐸1 𝐸2 𝐸3 ]. Four
fall prediction system. The prediction time (latency) will remain different configurations of frame designs are investigated to
the same as 300-700msec due to the utilization of parallel find the ideal frame length that achieves the optimum trade-off
computation of all features in the proposed seven features between detection accuracy, complexity and system latency.
implementation. Fig. 6 summarizes the comparison for different windowing
configurations. These frame designs are described as 1) a 2 sec
Table II: Resources and accuracy comparison between single feature FMFP non-overlapped frame; 2) a 1 sec non-overlapped sliding frame;
and the proposed 7 features FMFP. 3) a 2 sec overlapped frame with 0.5 sec augmentation, and 4)
a 1 sec overlapped frame with 0.5 sec augmentation. Analytical
results demonstrate that case 1) has the lowest hardware
complexity but with low performance (detection accuracy of
75% and the highest system latency (2 sec)). Both cases of 3)
and 4) have small latency (0.5 sec) and medium performance
(detection accuracy of 85% and 87%, respectively) but with a
huge cost for hardware implementation. Whereas, case 2)
achieves medium latency (1 sec), and high performance
(detection accuracy of >95%) with relaxed hardware
implementation cost. The FV evaluation through the non-
overlapping frame will result in reduced hardware resources
V. PROPOSED PS-THRESHOLD BASED FALL DETECTION compared to the overlapping one, for example, 2) will consume
ALGORITHM 27% fewer resources compared to 4) due to a) increase in the
storage requirement and b) enhanced clock frequency for the
For distinguishing a fall event from an ADL, a fall event is
overlapping data. Therefore, a frame size of 1 sec is selected in
divided into three stages, i.e., the pre-fall phase, the fall impact,
our proposed design and the FV is updated after each 1 sec with
and the post-fall phase. The FV formation is formed based on
the new incoming acceleration data. The fall event is also
tracking the SVS values that occur in 3 consecutive frames,
detected after 1 sec from its occurrence.
where 𝐸𝑖𝑗 is an indication of the 𝑆𝑉𝑆 value in each frame, ‘i’ is
Most of the conventional systems identify a fall as having
the frame number 1, 2 or 3 and ‘j’ is the iteration number, shown higher acceleration than ADLs that exceeds a threshold value
in (5). [25]. However, it is unreasonable to consider a constant
𝑋𝑗 = [ 𝐸1𝑗 𝐸2𝑗 𝐸3𝑗 ] (10) acceleration value as a threshold above which all motions are
considered as a fall and vice versa. Because this value may
Where 𝑋𝑗 is the FV. These FVs will be used to identify the differ for people in different situations, health conditions and
fall events. They are explained as follows: ages [32]. Selecting a low threshold may increase false
A. Pre-Fall Phase: A fall normally occurs when a person is positives during ADL, resulting in lower specificity. On the
carrying out a regular movement such as walking or sitting contrary, too stringent threshold selection may lead to failure of
which can be labeled by low to medium acceleration values that
will mark the beginning of a fall case. During this phase, the 𝑆𝑉𝑆
1534-4320 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TNSRE.2019.2911602, IEEE
Transactions on Neural Systems and Rehabilitation Engineering
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
Fig. 6: Different confiurations for sliding frame FV formation for fall detection (a) 2 sec non-overlapped frame, (b) 1 sec non-overlapped frame, (c) 2 sec
frame with 0.5 sec increment and, (d) 1 sec frame with 0.5 sec increment.
the fall detection algorithm. Fig. 7 shows the SVS for two
patients for a fall event (front fall on knees) and an ADL case
(ascending stairs). Subject A is 30 years old male with a height
of 177 cm and weight of 102 Kg, while Subject B is a 26 years
old female with a height of 170 cm and weight of 90 Kg. From
Fig. 7 (a) and (c), Subject A needs to have a high PS_TH to
detect fall case correctly (~1.0), while Subject B in Fig. 7 (b)
and (d) needs a smaller value of PS_TH (~0.75). Selecting the
same threshold value for both subjects will erroneously detect
an ADL as a fall or vice versa. The PS_TH value is determined
once for each patient off-chip. The PS threshold (PS_TH) is
computed off-line with the help of MATLAB and uploaded
through SPI protocol to the FPGA. In this work, linear
regression is utilized to determine the optimal PS_TH to
perform on-sensor processing and reduce the Bluetooth data
overhead. Each subject’s PS_TH need to be determined,
therefore, the training set is required.
The flow chart of the proposed fall detection algorithm is
presented in Fig.8. Fall detection decision formation is based on
the 3-bits FV formed by the 3 consecutive “1-sec” frames. A fall
event is identified by a ‘010’ code as it resembles a high
Fig. 7: SVS for two subjects for a fall and no fall cases (a) and (b) are the
amplitude in the central frame only. For the cases where the last fall cases for patients A and B, respectively, (c) and (d) are the ADL case for
two bits are ‘01’, they can be fall events. Hence, the verdict is patients A and B, respectively.
delayed until the next cycle after receiving the adjacent 1-sec
time frame of the acceleration data. In rare cases where a fall
VI. MEASUREMENT RESULTS AND DISCUSSION
event produces two spikes larger than the PS_TH value in two
consecutive 1-sec frames, the SMFD system will check if there To verify the working of the proposed system in the real
was a fall event predicated by FMFP to decide a fall or no fall environment, it was tested on elder subjects. Despite the non-
case. The coordination between the two modes of the systems invasiveness nature of the proposed system, the Institutional
will make the overall system more reliable and reduce false Review Board (IRB) approval was processed along with the
negatives. The remaining cases are considered ADLs since the participant consent form under the IRB Organization number
conditions for fall event are not met. If a fall event is identified, LUMS-IRB20170114 for the device testing in human subjects.
the tri-axial acceleration information is kept in the local All the falls and ADLs are performed in a realistic way similar
repository and an alarming signal is transmitted to the health to the method utilized in MobiFall dataset [27]. However, we
care provider through the internet. The fall will be spotted after used a thicker mattress (10 cm) to dampen the fall and protect
1 sec of the occurrence which is >50% less latency than the elder subjects who participated in the experiments. The
[1],[2],[23]. Fall/ADL validation experiments are performed by 20 different
subjects (12 males and 8 females) (age: 65-70) (Body Mass
1534-4320 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TNSRE.2019.2911602, IEEE
Transactions on Neural Systems and Rehabilitation Engineering
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
Index (BMI): 28.3-35.1). The validation experiments combined Fig. 10 (a) shows the logic resources utilized by the proposed
with the MobiFall dataset comprise of 6 different fall cases and SMFD/FMFP system (feature extraction and classification) on
11 multiple ADLs from a total of 77 subjects with greater than FPGA Virtex5 (XC5VLX110T), which result into equivalent
600 recorded trials. The subjects in the MobiFall dataset are logic cells of 62.3K [39]. Power budget distribution of the
healthy 42 males (age: 20-47 years, BMI: 19.78-34.68) and 15 sensor, microcontroller, and FPGA for the proposed system is
females (age: 20- 36 years, BMI: 18.36-31.14) [27]. The ADLs shown in Fig. 10 (b). The power distribution chart focuses
are selected based on their consistency and resemblance to mainly on processing power with SMFD/FMFP consumes
realistic falls that may increase the false positives. Explicitly, 0.34mW at an operating clock of 1 KHz, and does not include
they are chosen established on the subsequent norms: a) fall-like wireless transmission power, which will dominate the power
events where the subject typically stays immobile at the end, for budget.
example, stepping in a car or sitting on a seat; b) ADL like
standing, walking, and descending and ascending the stairs; c)
abrupt and rapid events that are alike to fall events, such as
jogging and jumping. The four falls comprise a) forward lying
fall, b) back chair fall, c) front knees fall, and d) side fall.
99.1%
True Class
raised to 1 with pre-fall impact time of 400msec and 600msec, 99.3% 0 99.1% 0.9%
0 99.3% 0.7% 0.9%
respectively. From the offline training of the proposed SMFD 0.7%
system, PS_TH is computed to be 1.5 for this subject. Each case 98.6% 1 2.2% 97.8% 97.8%
1 1.4% 98.6%
1.4% 2.2%
in Fig.9 resulted in a unique 3-bit output FV that is passed to the 0 1
0 1
detection algorithm for fall detection. The proposed SMFD TPR/FNR Predicted Class TPR/FNR
algorithm correctly identifies jumping and sitting on a seat as No
(a) (b)
Fall and front knee- fall and falling on right sideways as fall
Fig. 11: Confusion Matrix of the proposed fall a) detection and b) prediction.
cases.
1534-4320 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TNSRE.2019.2911602, IEEE
Transactions on Neural Systems and Rehabilitation Engineering
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
1534-4320 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TNSRE.2019.2911602, IEEE
Transactions on Neural Systems and Rehabilitation Engineering
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
1534-4320 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.